Draft for put landing page; identified TODOs (#34)

* Draft for put landing page; identified TODOs Issue: #25 * Completed tf surgery; Identify all TODOs in golang (#35) * Complete tf surgery; Identify all TODOs in golang For #25 * fix compile error; progress in metadata cronjob add query * Ready to test (#36) * Ready to test * Fix db field first char not lowercase Tracked by #25 (comment) * Fix permission of db index, S3 pull Tracked by #25 (comment) * All tests complete Tracked by #25 (comment)
rivernews · Sep 29, 2022 · 0e1ee8f · 0e1ee8f
1 parent eb61eef
commit 0e1ee8f
Show file tree

Hide file tree

Showing 23 changed files with 646 additions and 329 deletions.
diff --git a/.gitignore b/.gitignore
@@ -1,10 +1,11 @@
 **credential**
 **/builds/**
 
-lambda_golang/landing
-lambda_golang/stories
-lambda_golang/landing_metadata
-lambda_golang/story
+lambda_golang/*
+!lambda_golang/go.mod
+!lambda_golang/go.sum
+!lambda_golang/*/
+!lambda_golang/*/**
 venv
 
 # Binaries for programs and plugins

diff --git a/cloud_environments/terraform.sh b/cloud_environments/terraform.sh
@@ -22,9 +22,10 @@ set +o allexport
 if (
     cd $GOLANG_SRC_DIR && \
     go build ./cmd/landing && \
-    go build ./cmd/landing_metadata && \
+    go build ./cmd/landing_metadata_cronjob && \
     go build ./cmd/stories && \
     go build ./cmd/story && \
+    go build ./cmd/stories_finalizer && \
     cd $PYTHON_SRC_DIR && python -m compileall layer src
 ); then
     cd $DEPLOY_DIR
@@ -37,7 +38,7 @@ if (
     # https://github.com/terraform-aws-modules/terraform-aws-step-functions/issues/20
     # terraform "$@" \
     #     -target=module.main.module.scraper_lambda \
-    #     -target=module.main.module.landing_parse_metadata_lambda
+    #     -target=module.main.module.landing_metadata_cronjob_lambda
 
     terraform "$@"
 else

diff --git a/cloud_module/dynamodb/table.tf b/cloud_module/dynamodb/table.tf
@@ -1,12 +1,12 @@
 resource "aws_ssm_parameter" "media_table" {
   name  = "/app/media-literacy/table"
   type  = "String"
-  value = aws_dynamodb_table.media_table.arn
+  value = "${aws_dynamodb_table.media_table.arn},${aws_dynamodb_table.media_table.id}"
 }
 
 // https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/dynamodb_table#attributes-reference
 resource "aws_dynamodb_table" "media_table" {
-  name           = "Mediatable"
+  name           = "${title(replace("${var.project_alias}_${var.environment_name}", "-", "_"))}"
   billing_mode   = "PROVISIONED"
   read_capacity  = 20
   write_capacity = 20
@@ -23,8 +23,12 @@ resource "aws_dynamodb_table" "media_table" {
     type = "S"
   }
 
+  attribute {
+    name = "s3Key"
+    type = "S"
+  }
+
   // other fields
-  // S3 key
   // docType = {landing | story | landingMetadata | ...}
   // events
 
@@ -58,6 +62,15 @@ resource "aws_dynamodb_table" "media_table" {
     non_key_attributes = ["s3Key"]
   }
 
+  global_secondary_index {
+    name               = "s3KeyIndex"
+    hash_key           = "s3Key"
+    range_key          = "createdAt"
+    write_capacity     = 10
+    read_capacity      = 10
+    projection_type    = "KEYS_ONLY"
+  }
+
   tags = {
     Project = local.project_name
     Environment = var.environment_name

diff --git a/cloud_module/pipeline/global_ssm.tf b/cloud_module/pipeline/global_ssm.tf
@@ -9,4 +9,8 @@ data aws_ssm_parameter media_table {
 locals {
     newssite_economy_tokens = split(",", data.aws_ssm_parameter.newssite_economy.value)
     newssite_economy_alias = local.newssite_economy_tokens[2]
+
+    _media_table_tokens = split(",", data.aws_ssm_parameter.media_table.value)
+    media_table_arn = local._media_table_tokens[0]
+    media_table_id = local._media_table_tokens[1]
 }
diff --git a/cloud_module/pipeline/lambda.tf b/cloud_module/pipeline/lambda.tf
@@ -59,7 +59,7 @@ module "step_function" {
 module "scraper_lambda" {
   source = "terraform-aws-modules/lambda/aws"
   create_function = true
-  function_name = "${local.project_name}-scraper-lambda"
+  function_name = "${local.project_name}-landing-lambda"
   description   = "Lambda function for scraping"
   handler       = "landing"
   runtime     = "go1.x"
@@ -82,6 +82,13 @@ module "scraper_lambda" {
 
   attach_policy_statements = true
   policy_statements = {
+    allow_db_query = {
+      effect    = "Allow",
+      actions   = [
+        "dynamodb:PutItem"
+      ],
+      resources = [local.media_table_arn]
+    }
     s3_archive_bucket = {
       effect    = "Allow",
       actions   = [
@@ -98,6 +105,7 @@ module "scraper_lambda" {
     S3_ARCHIVE_BUCKET = data.aws_s3_bucket.archive.id
 
     NEWSSITE_ECONOMY = data.aws_ssm_parameter.newssite_economy.value
+    DYNAMODB_TABLE_ID = local.media_table_id
   }
 
   tags = {

diff --git a/cloud_module/pipeline/landing_s3_trigger.tf b/cloud_module/pipeline/landing_s3_trigger.tf
diff --git a/cloud_module/pipeline/stories_sqs.tf → cloud_module/pipeline/s3_triggers.tf b/cloud_module/pipeline/stories_sqs.tf → cloud_module/pipeline/s3_triggers.tf
@@ -1,35 +1,31 @@
-module "stories_queue" {
-  source  = "terraform-aws-modules/sqs/aws"
-  version = ">= 2.0, < 3.0"
-
-  # SQS queue attributes: https://docs.aws.amazon.com/AWSSimpleQueueService/latest/APIReference/API_CreateQueue.html
-
-  # FIFO queue should append suffix .fifo
-  name = "${local.project_name}-stories-queue"
-
-  delay_seconds = 0
-
-  # so we can use per-message delay
-  fifo_queue = false
-
-  # FIFO queue only
-  # content_based_deduplication = true
-
-  visibility_timeout_seconds = 3600
+resource "aws_s3_bucket_notification" "bucket_notification" {
+  bucket = data.aws_s3_bucket.archive.id
+
+  lambda_function {
+    lambda_function_arn = module.landing_metadata_s3_trigger_lambda.lambda_function_arn
+    events              = ["s3:ObjectCreated:*"]
+    filter_prefix       = "${local.newssite_economy_alias}/"
+    filter_suffix       = "/metadata.json"
+  }
 
-  # enable long polling
-  receive_wait_time_seconds = 10
+  depends_on = [
+    aws_lambda_permission.allow_bucket_trigger_by_landing_metadata
+  ]
+}
 
-  tags = {
-    Project = local.project_name
-  }
+resource "aws_lambda_permission" "allow_bucket_trigger_by_landing_metadata" {
+  statement_id  = "AllowExecutionFromS3Bucket"
+  action        = "lambda:InvokeFunction"
+  function_name = module.landing_metadata_s3_trigger_lambda.lambda_function_arn
+  principal     = "s3.amazonaws.com"
+  source_arn    = data.aws_s3_bucket.archive.arn
 }
 
-module "stories_queue_consumer_lambda" {
+module "landing_metadata_s3_trigger_lambda" {
   source = "terraform-aws-modules/lambda/aws"
 
   create_function = true
-  function_name = "${local.project_name}-fetch-stories"
+  function_name = "${local.project_name}-stories-lambda"
   description   = "Fetch ${local.project_name} stories; triggered by metadata.json creation"
   handler       = "stories"
   runtime     = "go1.x"
@@ -62,30 +58,20 @@ module "stories_queue_consumer_lambda" {
 }
 EOF
 
-  # event source mapping for long polling
-  event_source_mapping = {
-    sqs = {
-      event_source_arn = module.stories_queue.this_sqs_queue_arn
-      batch_size = 1
-    }
-  }
-  allowed_triggers = {
-    sqs = {
-      principal  = "sqs.amazonaws.com"
-      source_arn = module.stories_queue.this_sqs_queue_arn
-    }
-  }
   attach_policy_statements = true
   policy_statements = {
-    pull_sqs = {
+    allow_db_put = {
       effect    = "Allow",
-      actions   = ["sqs:ReceiveMessage", "sqs:DeleteMessage", "sqs:GetQueueAttributes"],
-      resources = [module.stories_queue.this_sqs_queue_arn]
+      actions   = [
+        "dynamodb:UpdateItem",
+      ],
+      resources = [
+        local.media_table_arn,
+      ]
     }
     s3_archive_bucket = {
       effect    = "Allow",
       actions   = [
-        "s3:PutObject",
         "s3:GetObject"
       ],
       resources = [
@@ -107,8 +93,10 @@ EOF
     SLACK_WEBHOOK_URL = var.slack_post_webhook_url
     LOGLEVEL = "DEBUG"
     ENV = local.environment
+    DEBUG = "true"
 
     S3_ARCHIVE_BUCKET = data.aws_s3_bucket.archive.id
+    DYNAMODB_TABLE_ID = local.media_table_id
     SFN_ARN = module.batch_stories_sfn.state_machine_arn
   }
 

diff --git a/cloud_module/pipeline/scheduler.tf b/cloud_module/pipeline/scheduler.tf
@@ -60,3 +60,90 @@ data "aws_iam_policy_document" "scheduler" {
     }
   }
 }
+
+
+resource "aws_cloudwatch_event_rule" "landing_metadata_scheduler" {
+  count = var.environment_name == "" ? 1 : 0
+
+  name                = "${local.project_name}-schedule-start-metadata-for-landing"
+  # schedule experssion
+  # https://docs.aws.amazon.com/eventbridge/latest/userguide/scheduled-events.html
+  schedule_expression = "rate(1 hours)"
+  description         = "Every hour to give courtesy to the website"
+}
+
+resource "aws_cloudwatch_event_target" "landing_metadata_scheduler_event_target" {
+  count = var.environment_name == "" ? 1 : 0
+
+  target_id = "${local.project_name}-schedule-start-metadata-for-landing-event-target"
+  rule      = aws_cloudwatch_event_rule.landing_metadata_scheduler.0.name
+  arn       = module.landing_metadata_cronjob_lambda.lambda_function_arn
+}
+
+module landing_metadata_cronjob_lambda {
+  source = "terraform-aws-modules/lambda/aws"
+  create_function = true
+  function_name = "${local.project_name}-landing-metadata-cronjob-lambda"
+  description   = "Query landing pages in db; compute & archive their metadata"
+  handler       = "landing_metadata_cronjob"
+  runtime       = "go1.x"
+
+  source_path = [{
+    path = "${var.repo_dir}/lambda_golang/"
+    commands = ["${local.go_build_flags} go build ./cmd/landing_metadata_cronjob", ":zip"]
+    patterns = ["landing_metadata_cronjob"]
+  }]
+
+  timeout = 900
+  cloudwatch_logs_retention_in_days = 7
+
+  publish = true
+
+  attach_policy_statements = true
+  policy_statements = {
+    allow_db_query = {
+      effect    = "Allow",
+      actions   = [
+        "dynamodb:Query",
+        "dynamodb:UpdateItem",
+      ],
+      resources = [
+        local.media_table_arn,
+        "${local.media_table_arn}/index/metadataIndex"
+      ]
+    }
+    s3_archive_bucket = {
+      effect    = "Allow",
+      actions   = [
+        "s3:GetObject",
+        "s3:PutObject",
+      ],
+      resources = [
+        "${data.aws_s3_bucket.archive.arn}/*",
+      ]
+    }
+    # enable getting 404 instead of 403 in case of not found
+    # https://stackoverflow.com/a/19808954/9814131
+    s3_archive_bucket_check_404 = {
+      effect    = "Allow",
+      actions   = [
+        "s3:ListBucket",
+      ],
+      resources = [
+        "${data.aws_s3_bucket.archive.arn}",
+      ]
+    }
+  }
+
+  environment_variables = {
+    SLACK_WEBHOOK_URL = var.slack_post_webhook_url
+    LOG_LEVEL = "DEBUG"
+    DEBUG = "true"
+    S3_ARCHIVE_BUCKET = data.aws_s3_bucket.archive.id
+    DYNAMODB_TABLE_ID = local.media_table_id
+  }
+
+  tags = {
+    Project = local.project_name
+  }
+}
diff --git a/cloud_module/pipeline/sfn_def/batch_stories_def.json b/cloud_module/pipeline/sfn_def/batch_stories_def.json
@@ -12,6 +12,7 @@
       "Parameters": {
         "story.$": "$$.Map.Item.Value",
         "newsSiteAlias.$": "$.newsSiteAlias",
+        "landingPageUuid.$": "$.landingPageUuid",
         "landingPageTimeStamp.$": "$.landingPageTimeStamp"
       },
       "Iterator": {
@@ -32,6 +33,11 @@
           }
         }
       },
+      "Next": "Stories-Finalizer"
+    },
+    "Stories-Finalizer": {
+      "Type":"Task",
+      "Resource": "${STORIES_FINALIZER_LAMBDA_ARN}",
       "End": true
     }
   }